CSIRO Mineral Resources Discovery Internship Program

We are seeking applicants for CSIRO Mineral Resources Discovery Program's student

internship program, running between July 2020 and June 2021.

Applications are open for outstanding university students (3rd year undergraduate or postgraduate

level) to join CSIRO Mineral Resources for a 3-month internship to work on projects directly related

to the Discovery Program's two impact areas: Exploration Through Cover, and Orebody Knowledge.

CSIRO offers a total living allowance of $6000 to a minimum of 3 successful interns. Note that

upfront travel costs (including visas) will need to be met by the successful applicants. All internships

will be based in Perth, Western Australia, unless otherwise specified.

Available projects for 2020-2021:

• Experimental techniques in spectral geological measurements

• Student internship proposal: Preparation of electromagnetic (EM) system descriptions

• Characterisation of New Caledonian Nickel Laterite

• The role of trace Ta-Sn minerals in Lithium pegmatites as indicators of redox and pH

conditions

• Weathering effects on REE in deeply weathered terranes

• Mineral exploration of carbonatites: Australia and India

• Mineral characterization and data set integration to assess REE potential of carbonatites

• Development of an online QAQC and calibration tool for portable XRF analysers

• Thermodynamic modelling of subduction fluid related Au-Fe-S ore deposits

• Data assimilation of earthquake sequences to better understand giant mineralisation events

• Equation-of-state and dielectric constant of magmatic-hydrothermal NaCl fluid

• Deformation and Mineral Replacement at Jundee Gold Mine

• Are basement faults and veins linked to Pilbara gold nuggets?

• Creating a geological training dataset for machine learning from drill core photos

To apply for an internship please submit the following to Monica LeGras (monica.legras@csiro.au)

before the 3rd March 2020:

• Full CV, including current university course, expected completion dates, and contact details

of an academic supervisor.

• A cover letter, including a summary of your personal areas of research interest and how they

align with the two impact areas of the Discovery Program.

• Your preferred project title(s), including why you are interested in that project.

Note that all interns MUST be registered as a student at a university DURING their internship,

or provide evidence of enrolment in a university course that commences after the internship.

Interns will be expected to write a report or deliver an oral presentation communicating their research

at the completion of the internship.

Applicants will be advised of the outcome of their applications by the 31st of March 2020.

Experimental techniques in spectral geological measurements

Supervisor: Ian Lau Reflectance spectrometers are used for the rapid collection of mineralogy, which can be used for mapping alteration mineral footprints of hydrothermal ore deposits, ground validation of spaceborne and airborne spectral data sets and voluminous measurements on drill core (i.e. hyperspectral drill core logging). The impacts of the measurement technique on the results for this spectrometer has many uncertainties. This project would involve the undertaking of a range of spectral measurements, using the wide range of spectrometers at the disposal of CSIRO Mineral Resources in Kensington, Western Australia on different geological and standard materials which have undergone varying treatments. The materials consist of pure mineral samples from a range of exotic locations and world-class mineral deposits, which make up a spectral reflectance reference library, and calibrated laboratory standards. The data from the measurements will be published on an online spectral library of minerals and materials. The aim of the study is to improve the understanding and estimation of the uncertainty in spectral measurement techniques, with the results from this work used for the publication of guidelines in spectral measurements. The student will become familiar with the operation of different kinds of instrumentation and software and the processing of their subsequent data.

Preparation of electromagnetic (EM) system descriptions

Supervisors: Aaron Davis and Shane Mulè

Electromagnetics are often used as a primary geophysical technique for minerals exploration, groundwater detection and geohazard mapping. Owing to its relatively low cost, airborne electromagnetics (AEM) offers unprecedented spatial coverage of regions of interest for geophysical exploration. However, the data from AEM and other ground-based EM surveys needs to be processed and inverted into meaningful electrical conductivity-depth models to realise its full utility.

Uploading AEM data to processing and inversion software packages is mostly manual and constitutes

a labour-intensive part of post-survey geophysical exploration. The set-up cost of importing EM data

can be high due to the complexity of mapping survey data into useful parameters that are recognised

by the software. Although there is an industry standard available that contractors use to format data

into human-readable files, there is tremendous variability in EM systems, data structure, data names,

and geophysical units. Moreover, there is variability in the interpretation of the industry standard for

formatting data.

Efforts are currently being made to create a new standard of geophysical data based on the netCDF

format. Although this has the potential to streamline data import, it will not affect legacy and

historical data sets: there is still a need to be able to import old datasets in new software.

This project proposal attempts to address the problem of preparing EM data for a new standard of

importation. By scanning through description files of legacy and publicly available data, the researcher

will be responsible for generating a probabilistic import device that will convert old data into netCDF

format. Using word recognition and semantic relationship methods, the software import package will

significantly decrease set-up cost of AEM surveys and will contribute to a new industry standard for

geophysical electromagnetic data.

The import package will be directly linked to the new CSIRO Geophysical Processing Toolkit.

Supervision will be provided by a senior research scientist, and at least one publication is planned in

Exploration Geophysics. Presentation of the completed product at AEGC 2021 will be strongly

encouraged.

Characterisation of New Caledonian Nickel Laterite

Supervisor: Erick Ramanaidou The valuable Nickel deposts in New Caledonia are related to a complex Nickel laterite profile. These Nickel laterite profiles are composed of 6 weathering horizons including from the base to the top (1) the fresh rock either a dunite (olivine) or a harzburgite (olivine et pyroxenes) totally or partially serpentinised (lizardite, antigorite et chrysotile); (2) a saprolite; (3) a transition laterite with asbolane; (4) a yellow laterite; a red laterite and (6) a ferruginous duricrust. The project is aimed at understanding the detailed distribution of Ni and Co through the entire lateritic profile. A range of chemical (Minalyze, scanning electron microscope), mineralogical (XRD, reflectance and Raman spectroscopies) and textural methods (Medical and micro CT) will be used for the characterisation and distribution of Ni into the various minerals.

The role of trace Ta-Sn minerals in Lithium pegmatites as indicators of redox and pH conditions

Supervisor: Alex Otto Hardrock lithium deposits are associated with felsic pegmatites that are believed to have originated from late stage granite intrusions or anatectic melts of metasediments. These lithium pegmatites fall into two categories: (1) spodumene bearing, relatively homogenous intrusions, and (2) lepidolite ± spodumene bearing, mineralogically zoned bodies. In some deposits Ta is enriched to ore grades and is being mined as a by-product. Currently there is little understanding what governs the occurrence of these two types of lithium pegmatites and how to identify them during early exploration efforts. Previous LA-ICP-MS studies revealed oscillatory zonation of spodumene in its Fe and Ga content, pointing to variable redox conditions in these systems. Initial investigations of the trace mineralogy revealed a complex interplay between Ta, Nb, and Sn minerals potentially related to the melt-fluid evolution. Additionally, trace amounts of As and Ag minerals have been found, yet it is unexplained. The intern student will utilise samples from the spodumene bearing Goulamina lithium deposit in Mali. The planned work includes detailed SEM imaging and chemical mapping. The data will be integrated with existing MAIA mapper (XRF) images, LA-ICP-MS analysis and maps, and quantitative mineralogy and geochemistry.

Weathering effects on REE in deeply weathered terranes

Supervisor: Ignacio González-Álvarez

Critical metals are often referred to as the “oil of the twenty-first century” due to their growing importance in advanced technology, which spans from production of alternative energy, through the communications industry, to cutting edge technology. The understanding of REE and HFSE (Zr, Hf, Nb, Ta, Th, U) mobility and accumulation processes is fundamental to their exploration. REE and HFSE have a preferred trivalent oxidation state (with the exception of Ce and Eu, which are redox sensitive and can have both 2+ and 4+ forms). The original REE budget of rocks can be mobilized, dispersed, fractionated or accumulated under specific conditions such as during weathering, hydrothermal fluid flow or basinal brine activity. Therefore, REE can be used as an efficient proxy to provide insights into geochemical dispersion processes in regolith and sedimentary sequences (Fig. 1). This project will build on existing geochemical and mineralogical datasets from diverse geological contexts to assess how intense weathering fractionates and accumulates REE, and how this knowledge can be used to provide a vector for mineral exploration.

Figure 1. Example of a REE study in a regolith/sedimentary sequence, South Australia (González-

Álvarez et al. 2018).

Mineral exploration of carbonatites: Australia and India

Supervisor: Ignacio González-Álvarez

Western Australia and India share common mineral exploration challenges since both countries display extensive areas covered by sedimentary sequences and/or deeply weathered profiles. The detection of geochemical footprints of carbonatites in this geological context is a significant challenge facing mineral exploration of critical metals in Australia and India. Rare Earth Elements (REE) and High Field Strength Elements (HFSE; Nb, Ta, Zr, Hf, Th, U) are abundant in Earth's crust. Nevertheless, known economic deposits of such are fewer than for most other ores. Carbonatites are defined as igneous rocks composed of >50 wt% primary igneous carbonate minerals and <20wt.% SiO2. High REE and HFSE concentrations are distinctive of carbonatite rock suites and alkaline complexes, which are the most critical rock reservoir of REEs and other critical elements. This project seeks to elucidate a better understanding of diverse approaches to the exploration for carbonatites under cover, and will compile and compare mineral exploration proxies that have been efficiently used in Australia and India.

Figure 1. Main carbonatite deposits of Nb and REE in India and Australia.

Mineral characterization and data set integration to assess REE potential of carbonatites

Supervisors: Siyu Hu and Ignacio González-Álvarez

The demand for rare earth elements (REE) is increasing, and the exploration of REE-related deposits in Australia is prioritized as part of the critical elements Strategy. Carbonatites represent some of the rarest lithologies on Earth, and one of the richest in essential elements such as REE. There are ~530 carbonatites that have been globally identified, and two of them are currently mined for REE in Western Australia. But research access to complete and diverse data sets on weathered carbonatites in Australia is limited. However, Australian companies that explore in Africa have collected data on carbonatites that have experienced similar weathering conditions to Australia. Hence this information, already extant, is of crucial importance in improving carbonatite exploration protocols in Australia. Meanwhile, technology has made possible the rapid collection of a wide variety of datasets for mineral exploration programs. Digital and analytical technologies have provided the tools to be able to analyze and integrate diverse datasets and large volumes of data to provide critical information on the key parameters controlling the formation of ore deposits, along with assisting and directing the mineral exploration program. This new data-integration context is essential in better evaluating the potential of exploration targets and to implement more efficient mineral exploration protocols. An Australian company (Kinloch Resources Ltd.) has provided CSIRO with a wide variety of datasets from Ondurakorume, Namibia. These datasets span geology, tectonics, geochemistry, remote sensing, mineralogy, XRD, SEM, TIMA, and geophysics. This project will contribute to the characterization of the Ondurakorume carbonatite, and assess its REE potential, by integrating all available datasets. The outputs will be tailored to the Australian mineral exploration context.

Figure 1. Overview of the landscape associated with the Ondurakorume carbonatite, Namibia. Pictures courtesy of Kinloch Resources Ltd. (Exploration asset now under Osino Resource Corp.)

Development of an online QAQC and calibration tool for portable XRF analysers

Supervisors: Margaux Le Vaillant, Sam Bradley, Renee Birchall

Portable X-Ray Fluorescence (pXRF) analysers have become a widely used analytical tool within the

Minerals industry. Their performance keeps on increasing and the quality of the obtained geochemical

data with these rapid - in the field - data collection tools has improved over the past ten years.

However, there still does not exists an easy tool to help explorers who use the best practices by

collecting standards, monitors and blanks, to quickly evaluate the quality of their data and calibrate

it. Especially for field campaigns that require the use of multiple instruments and dataset need to be

calibrated back to the same standards for the data to be comparable.

The Discovery Program has already started developing an in-house tool to automate calibration of

pXRF data. With this internship, we aim at testing and developing the tool further. The produced

online tool would allow the user to quickly and easily assess the results of analyses obtained on blanks

and monitor, and use the analyses collected on standards to calibrate the data. This would reduce the

amount of time geologist spend processing pXRF data and combined with best practice workshops

help the exploration industry extract as much valuable information as possible from their pXRF

datasets.

This internship project would be divided in three parts:

1) Proceed to a data quality comparison for different instruments currently used by the Minerals industry, and evaluate the need for calibration in function of the instrument being used.

2) Tests different standard types (powders, pressed pallets, fused disks) to improve the calibration process and define best practices depending on the sample medium.

3) If time allows, develop the data Quality Assurance / Quality Control (QAQC) visualisation tool as a user-friendly interface to be integrated with the online processing toolkit.

This project would be supervised by Dr. Margaux Le Vaillant, geologist/geochemists and Mr. Sam

Bradley, engineer with expertise in software development. Dr. Le Vaillant will supervise the intern

while working on collection of test datasets, interpreting and assessing the quality of obtained

calibrated data, in function of the instrument and the standards used, as well as testing of the already

developed calibration tools. In a second stage of the project, the student will then work with Sam

Bradley in improving the robustness of the automated calibration tool.

Key interest and capabilities of the student: Data Analysis, Geochemistry, Instrument testing and

potentially Python programming.

Thermodynamic modeling of subduction fluid related Au-Fe-S ore deposits

Supervisor: Fang Huang

Different types of ore deposits, including Cu, Mo, Au, Pb, Zn etc., have been found in orogenic belts all over the world [1,2]. We know that orogenic belts are from plate collisions and subductions. During subduction, fluids coming out of subducted slabs can carry large amounts of metal complexes and anions to provide sources for later mineralization and ore formation. However, although many lab and field work has been done to investigate such processes, thermodynamic modeling is lacking. Due to limited knowledge of dielectric constants and density of water at high P-T, aqueous fluid thermodynamic modeling has long been constrained to P-T less than 5 kb and 600 C. Recent advances in Deep Earth Water (DEW) model [3] has enabled the thermodynamic modeling of fluid-rock interactions at elevated pressure and temperature conditions up to 6 GPa and 1200 C. Here I propose to use DEW model to study the Fe-Au-S system in subduction-zone to investigate related ore-formation process, because the dissolution and reprecipitation of iron sulfides in a gold rich fluid could result in gold substitution in the mineral lattice and is ubiquitously found in gold mineralization systems [4,5]. Sulfur has also been suggested to be an important anion to bind Au during ore formation [6]. A recent paper [7] has covered all major S species in the subduction-zone fluids. The intern will follow this work to add Au- and Fe-S complexes into the thermodynamic database and perform DEW calculation along the subduction-zone geotherm. By combining model results with previous work, we will have probably a better understanding of orogenic belt related Fe-Au ore deposits. [1] Hou, Z. Q., Song, Y. C., Li, Z., Wang, Z. L., Yang, Z. M., Yang, Z., ... & Wang, F. (2008). Thrust-controlled, sediments-hosted Pb-Zn-Ag-Cu deposits in eastern and northern margins of Tibetan orogenic belt: Geological features and tectonic model. MINERAL DEPOSITS-BEIJING-, 27(2), 123. [2] Ge, W., Wu, F., Zhou, C., & Zhang, J. (2007). Porphyry Cu-Mo deposits in the eastern Xing’an-Mongolian Orogenic Belt: Mineralization ages and their geodynamic implications. Chinese Science Bulletin, 52(24), 3416-3427. [3] Sverjensky, D. A., Harrison, B., & Azzolini, D. (2014). Water in the deep Earth: the dielectric constant and the solubilities of quartz and corundum to 60 kb and 1200 C. Geochimica et Cosmochimica Acta, 129, 125-145. [4] Rottier B., Kouzmanov K., Walle M., Bendezu R. and Fontbote L. (2016) Sulfide replacement processes revealed by textural and LA-ICP-MS trace element analyses: example from the early mineralization stages at Cerro de Pasco, Peru. Econ. Geol. 111, 1347–1367. [5] Wu, Y. F., Evans, K., Li, J. W., Fougerouse, D., Large, R. R., & Guagliardo, P. (2019). Metal remobilization and ore-fluid perturbation during episodic replacement of auriferous pyrite from an epizonal orogenic gold deposit. Geochimica et Cosmochimica Acta, 245, 98-117. [6] Pokrovski, G. S., Kokh, M. A., Guillaume, D., Borisova, A. Y., Gisquet, P., Hazemann, J. L., ... & Haigis, V. (2015). Sulfur radical species form gold deposits on Earth. Proceedings of the National Academy of Sciences, 112(44), 13484-13489. [7] Li JL, Schwarzenbach EM, John T, Ague JJ, Huang F, Gao J, Klemd R, Whitehouse MJ, Wang XS. Uncovering and quantifying the subduction zone sulfur cycle from the slab perspective. Nature Communications. 2020 Jan 24;11(1):1-2.

Data assimilation of earthquake sequences to better understand giant mineralisation events

Supervisor: Thomas Poulet

Episodic Tremor and Slip (ETS) events in subduction environments have been identified as the temporal signature of characteristic deep Earth processes also responsible for episodic thrust reactivation events. Due to extreme localisation of the chemical decomposition and fluid generation involved, such events are accompanied by high temperature high pore pressure fluid pulses which we hypothesise to play a major role in the formation of giant ore deposits. This project aims at calibrating a potential physical model responsible for such events using a data assimilation approach on available GPS signals in New Zealand. The validation of this physical model on ETS signals would confirm the theory and present a new critical mechanism for mineralisation, opening the door to new mineral exploration approaches.

Equation-of-state and dielectric constant of magmatic-hydrothermal NaCl fluid

Supervisors: Yuan Mei, Matthew Josh and Weihua Liu

Sodium chloride (NaCl) is major component in many types of hydrothermal fluids. Equation-of-state

(EOS) and dielectric constant (DC) of NaCl solution under magmatic-hydrothermal conditions is

important for understanding ion association and metal complexation in ore-forming fluids, as well as

estimating oil content in porous reservoir rock in oil-gas exploration. We are developing a cross-BU

collaborative project to combine experimental measurement of dielectric constant in our in-house

laboratory with classical molecular dynamics simulations to investigate the EOS and DC of NaCl brines

as functions of temperature, pressure and salinity. This project will provide essential data of NaCl

fluids, useful in wide range of theoretical, experimental, and modelling studies of elemental behaviour

in high P-T fluids relevant to mineral and energy systems.

This project is a collaboration between CMR Discovery Program, Deep Earth Imaging Future Science

Platform and CSIRO Energy. As a part of the project, the measurement of dielectric constants would

make a perfect internship project because the research could be conducted and results delivered in

three-month’s period. This project will be an excellent opportunity for the potential student to get

exposure to the cutting-edge technologies and apply them to mineral and oil-gas industry problems.

The successful candidate will work in the DC lab to perform DC experiments, and compare with the

MD simulation data, then apply those fluid properties to hydrothermal ore fluid chemistry. We expect

a potential publication of the research outcomes and will encourage successful applicant to consider

a joint PhD project with CSIRO and a collaborated University.

Deformation and Mineral Replacement at Jundee Gold Mine

Supervisors: Mark Pearce, Renee Birchall

Deformation and fluid-rock reactions are key processes in hydrothermal gold deposits. These two processes feedback resulting in localisation and if we are lucky a high-grade gold deposit. Jundee carries exceptionally high-grade mineralisation and recent characterisation of the mineralogy and chemistry by CSIRO has shown that early, somewhat cryptic fabrics are overprinted by later, hydrothermal alteration that hosts gold. Despite being located within fault zones, crystallographic fabrics in the gold lodes suggest very little shearing and that mineralisation occurs somewhat statically, thus questioning the paradigm that deformation plays a key role in gold mineralisation. This project will focus on incipient replacement textures, examining both the chemical alteration, crystal plastic deformation and brittle fracturing that accompany fault formation in the greenstones that host Jundee. The intern should be interested in both geochemistry and structural geology and will gain experience of both microchemical and microstructural analysis on samples from Jundee. The outputs will be a new understanding of the relative importance of reactions and deformation processes, which will act to guide exploration in similar greenstone terranes.

Gold veins from the mine walls at Jundee – how do these really form?

Are basement faults and veins linked to Pilbara gold nuggets?

Supervisors: Sam Spinks, Mark Pearce, Jessica Stromberg Gold nuggets in the Archean conglomerates of the Pilbara have been linked to the world-class Wits deposits in South Africa. The origin of the nuggets is poorly understood and new work from CSIRO has shown that there is at least some post depositional modification by hydrothermal fluids. Hydrothermal fluid flow through the host rocks is not obvious but using large-scale X-ray mapping subtle alteration halos on clasts can be observed. A 1.6 km deep drill core recently drilled by Artemis Resources provides another piece in the 2.8 billion year old puzzle. The drill core taps the basement beneath the Fortescue Group, which hosts the conglomerates. The basement of this core dated at ~2.9 Ga has been cut by veins and minor faults that indicate fluid flow through these rocks. This project will use multiscale chemical and microstructural analysis to examine if these faults are the conduits for the fluid flow that affected the gold nuggets. The outputs will shed light on the most poorly understood nuggets in Australia that are the subject of a modern-day gold rush.

A good day’s prospecting in the Heart of the Pilbara

Creating a geological training dataset for machine learning from drill core photos

Supervisors: June Hill and Umer Javeed

Note that this project will be based in Sydney, NSW, or Canberra, ACT (depending on the timing)

In mineral exploration, rapid and efficient drill core logging plays a vital role. This information is used

for making key decisions regarding planning and analysis of the drilling program. A wide range of

analytical devices are being used to acquire different types of data from drill core. These analytical

devices usually include the collection of high-quality images including both visible (RGB) images and

hyperspectral images. We are interested in extracting meaningful geological information from RGB

photos of drill core using machine learning. A good machine learning model requires a very large

library of high-quality labelled images for training purposes. However, creating, maintaining and

storing this kind of training data is an expensive, difficult and a time-consuming task. It involves an

expert geologist to label the key information from a core sample. A cloud based automated labelling

workflow, which can be easily accessed by the geological community, could provide the basis for

establishing a substantial library of training images.

We are looking for a person who can research and test potential cloud-based solutions for labelling

sets of core images. They would set up a workflow and trial the system. The suitable candidate would

have a good knowledge and experience of cloud computing platform (Azure, IBM and AWS). The

suitable candidate would need a good software development background. Preferably some

experience in the geosciences, although not an essential requirement.

During their tenure, the intern will:

• gain knowledge of the state-of-the-art in cloud computing platforms

• develop knowledge in machine learning, including convolution neural networks and

assembling training data sets

• develop skills in developing practical computer-based workflows for scientists

• develop scientific writing skills